Potentiometer Placement

[WizenedEE]

How do you all get your potentiometers connected to the shaft it's supposed to be measuring? We have a wrist that will only move about 2π/3 radians, with a 1/2 inch shaft sticking out of the end. Are there special potentiometers with a hole in the middle, like an encoder? I saw one thread where people were connecting the end of the potentiometer to some surgical tubing and the other end of the tubing to the shaft, and that looked cool, but inaccurate. Are there any other ways to do it?

A picture of our wrist is here: http://team3238.com/albums/wk4-2011/1/10/ Where the part that says "look out" (not sure why) is the part that rotates with the claw itself.

Thanks for the help!

[ttakashima]

Bore a hole slightly smaller in the end of the 1/2 shaft and press fit the pot's shaft in. Or if there is a flat head on the output from the encoder you can bore a hole the same diameter as the Pot's output shaft and drill/tap a set screw hole through the 1/2 shaft.

[MrForbes]

we seem to end up making a little arm to clamp onto the shaft, and attache the end of the arm to the robot part that moves. This year we discovered the magic of making the little arm of plywood. Seems to work fine, but this one is going to be replaced with a longer one that will clear the arm support plate.

The pot housing is mounted to a bent plate of aluminum. There is no connection between the arm pivot bolt and the pot.

[anthonyttu]

We have used the surgical tubing with the zip tie method in past years with very surprising results. It is very accurate with as long as they are close together (under an inch apart). The surgical tubing also gave some much needed dampening. We used it until we switched to optical encoders last year.

[SuperBK]

I like it Jim - thanks.

[Manoel]

Originally Posted by anthonyttu

We have used the surgical tubing with the zip tie method in past years with very surprising results. It is very accurate with as long as they are close together (under an inch apart). The surgical tubing also gave some much needed dampening. We used it until we switched to optical encoders last year.

Even better than surgical tubing is a small piece of pneumatic tubing

[Bochek]

Here is 2200's solution, First off, the axle that the arm rotates on is live, (meaning it spins with the arm) The axle is a 3/8 bolt just like the ones provided in the KOP. We modify them by turning down the last 1/2 inch of the bolt to .250" diameter.

Thinking ahead, we ordered some free belts and sprockets from Gates. The rest is a piece of aluminum, 2 #10 machine screws, a .250 ID .375 OD bearing, and a long shaft potetiometer.

PIC:

- Bochek

[Al Skierkiewicz]

Adam,
Most hobby stores, robot outlets (Banebots) and surplus stores have little nylon gears for sale. If you choose the right ratio, mount one gear on the pot shaft and the other gear on the arm section that rotates, you will have an accurate method of reading arm position. The rotary encoders that are used for AM transmissions are great for this application since they are continuous rotation.

[MrForbes]

Neat idea Al....although making small gears mesh requires that the parts be in alingment, and stay in alignment. That won't happen with the arm we built!

But adding a spring loaded tensioner to keep the gears meshed might help the situation.